Challenge teams tackle a range of interesting problems to solve. The most successful projects address a topic that holds great interest for the team. In recent years, ideas for projects have come from Astronomy, Geology, Physics, Ecology, Mathematics, Economics, Sociology, and Computer Science. It is very important that the problem a team chooses is what we call "real world" and not imaginary. A "real world" problem has measurable components. We use the term Computational Science to refer to science problems that we wish to solve and explain using computer models.
Here are some sample questions teams might wish to answer: Is our county going to run out of water? How are attitudes towards self-care in the prevention of disease changed? What is the liklihood that the deer population in Bandelier will run out of grassland? How can data sensors improve our lives?
Projects fall into a couple of categories. In one category are problems that have clear mathematical models with well-defined variables. For some examples of these, you can look at the Population Model (ppt) and Population Plot A (xls) files, A Model for Computational Science Investigations (ppt) and Falling Rock Model (Excel).
These are problems that are usually modelled using Java or Excel.
Other projects study what are called complex systems and examine emergent behavior or the behavior of a system based on the way components of a decentralized system interact with each other. Each individual follows rules that describe its behavior and the results show the outcomes based on the interactions between the different groups of individuals. These kinds of projects are modeled using StarLogo or NetLogo. "With StarLogo, you can model (and gain insights into) many real-life phenomena, such as bird flocks, traffic jams, ant colonies, and market economies." (Adventures in Modeling: Exploring Complex Dynamic Systems with StarLogo)
Here is a link to an article about a huge demonstration in Leipzig before the fall of the Berlin Wall that was not organized by a central authority. How did all of these people decide to come together on that particular day?http://www.sciencedaily.com/releases/2004/08/040805090440.htm
Learn about StarLogo NOVA (it's free) at http://education.mit.edu/.
Learn about NetLogo (it's free) at https://ccl.northwestern.edu/netlogo/
After a team has found an idea that is interesting, the members begin the process of focusing on the key questions they wish to examine. Mentors are very useful at this point because they help teams clarify and define precisely the questions to be answered. Often interesting problems are very large and it is essential to think about the parts that make up the whole of the problem. When the parts have been identified, then teams can decide where they wish to begin. Later, methods for solving other parts of the problem can be designed. Have a look at these links about breaking a problem into essential components:
A team may decide to work on a project a second year. The demonstrated level (quantity and quality) of work invested in the research, modeling, and implementation of a continued project must be comparable to that invested in a new project, for it to be judged competitively.
The best preparation for a follow on project and potential publication is to do some journal research to see what has been done before. Then do some twist on the work to make it unique such as doing a model in an agent based software instead of the languages commonly used in the articles. Then you could add in a behavioral factor that would be difficult to do in the traditional approaches, highlighting the advantages of the new approach.